People * David Idea Exploring memory-efficient fine-tuning techniques for improving Hawaiian-to-English translation using Apple's MLX framework, comparing multiple approaches and optimizing for Mac hardware. Details * Successfully fine-tuned gemma-3-4b-it-4bit on Mac M1 Ultra (128GB RAM) achieving 0.8296 semantic similarity score, a 3.6% improvement over the base model * Discovered
People: * David Idea: * Comparing fine-tuning performance on MacBook M3 Max, Mac Studio M1 Ultra, and Nvidia 4090 using MLX and Unsloth Details: * Tested fine-tuning Phi-3-mini-4k-instruct model * Followed this Jan 2025 MLX guide for Apple hardware * Used Unsloth library for Nvidia GPU * Dataset had 627 examples and used 500 training steps
People David Idea Exploring cloud GPU performance, cost, and usability for running open-source AI models (in comparison to local hardware and in context of recent learning on software to handle concurrent users) Details * Compared RunPod serverless and pods using Nvidia (vLLM) and AMD (sglang) * Benchmarked Nvidia 3090, 4090, RTX 6000
People: * David Idea: * Comparing two open-source projects, Onyx and Morphik, for self-hosted semantic search (and deep research) across Google Drive files (PDFs, Docs, Sheets, and Slides) Details: * Onyx: Great text-based search, flexible embedding models * Morphik: Powerful multi-modal search (text, images, graphs) * Onyx easily connects and syncs with Google Drive * Morphik
A Real-World Test of Prompt-to-UI AI Creating a usable interface for a brand-new app often eats up early project time. To see how far AI can shorten that stage, we pointed Magicpatterns.com at a brand idea called Po‘owai, a cash-flow tool for non-profits that have to juggle restricted
Bringing the Ho‘omo‘o App Home — No Tech Jargon Required You played with Ho‘omo‘o in Bolt.ai’s browser editor and now you’d like to run it on your own computer and save it on GitHub for safekeeping. Good news: it’s basically a three-minute job.
* People * David * Idea * For a given set of local LLM models that a set of users (or community) wants to use - how do you measure performance/UX on ways to serve it (based on ease of use, speed of processing and generation, level of accuracy, and features like multi-user-context
* People * David * Idea * For a given set of local LLM models that a set of users (or community) wants to use - what is the best/easiest way to serve it (based on ease of use, speed of processing and generation, and features like multi-context and * Details * Engines like llama.
* People * David * Idea * Use LangGraph workflows and nodes to collect and process QB export data in comparison to Po'owai DB export rows to generate annotated matching ids (focused on people, funding sources, and projects) * Details * LangGraph agentic workflows * Node-based Quickbooks Desktop Web Connector client * LangChain adapters for reasoning
Ref Link: Ola Participant Handbook Theme Between the first and second sessions, leaders will deepen their leadership journey with an executive coach before embarking on a transformative exploration in the second gathering, drawing inspiration from the saga of Hiʻiakaikapoliopele and using traditional and modern mixed art methods to visually articulate
Ref Link: Ola Participant Handbook Theme ʻO nā hōkū nō nā kiu o ka lani—The stars are the spies of heaven! This immersive experience, blending land and water, allows leaders to learn both practical and metaphorical lessons from kānaka who sailed on Hōkūleʻa and Hikianalia, honing evaluative, analytical, and
Ref Link: Ola Participant Handbook Theme LEADING SELF: UNEARTHING LAYERS OF OUR IDENTITY THRU KINOLAU Participants will embark on a leadership journey by sharing artifacts that represent their identity, making lau lau to explore leadership connections, and immersing in the stories of Nāonealaʻa and Laʻamaikahiki to deepen their sense of